ai-reliability
43 pieces
Field Guide to AI, Security and Cybercrime
Chatbot Memory: What Your AI Assistant Keeps About You
Your assistant keeps three separate records about you. Deleting one does not delete the others.
Opinion · 6 min
Deepfakes Enter the Courtroom: A Challenge for Evidence
The established rules for authenticating video and audio evidence are breaking down in the face of AI-generated fabrication.
Opinion · 9 min
Encrypted Search: What You Give Up for Privacy
End-to-end encryption protects your queries but fundamentally limits what a search service can do for you.
Opinion · 7 min
How AI Agents Turn Bad Sources Into Confident Actions
An agent does not just repeat misinformation; it converts it into actions and documents that carry your organisation's authority.
Opinion · 7 min
How to Check an AI System for Bias Before It Decides About People
Bias is found by testing outcomes across groups and examining the path from data to decision, not by reading a vendor's fairness statement.
Opinion · 9 min
AI Homework Help: When Faster Means Less Learned
Handing the struggle to a chatbot removes the part of homework that does the teaching.
Opinion · 10 min
AI in the Security Operations Center: What It Actually Speeds Up
Summaries and triage help analysts; autonomous response amplifies every false positive.
Opinion · 9 min
AI Productivity Claims: Faster Is Not the Same as Better
Time saved producing a draft often reappears as time spent checking it.
Opinion · 8 min
AI Red Teaming for Small Teams: A One-Week Plan
Test what your AI features can do when fed hostile content, not how rude a jailbreak can make them.
Opinion · 9 min
AI Slop and Fake Reviews: Finding Human Signal Online
When writing is free to produce, the evidence of a real experience is what is scarce.
Opinion · 7 min
AI Sycophancy: Why Chatbots Agree With You, and Why It Matters
Agreement is trained in by human preference, and it quietly bends judgement.
Opinion · 9 min
Can AI Write Malware? What Has Actually Changed for Attackers
Language models cut the cost of variation and lures far more than they invent new attacks.
Opinion · 9 min
Content Credentials (C2PA) Explained: What the Label Proves
Provenance can vouch for what honest sources signed; its absence proves nothing about anything else.
Opinion · 10 min
Fake Citations: How AI Hallucinations Reach Courts and Journals
The failure is not the model inventing sources; it is institutions that never check references.
Opinion · 9 min
Fake Customer Support Numbers: The Scam That Starts in Search
When you search for a help line, the results page itself can be the phishing channel.
Opinion · 7 min
How to Check a Viral Video Is Real Before You Share It
Pixel-peeping loses; tracing who posted it first, when, and where usually wins.
Opinion · 9 min
Should You Trust AI Medical Advice? How to Use It Safely
Chatbots are good at explaining and preparing questions, and keeping decisions where accountability exists.
Opinion · 10 min
Slopsquatting: When AI Invents a Package, Someone Registers It
Hallucinated dependency names repeat across users and sessions, which makes them predictable targets.
Opinion · 10 min
Using ChatGPT as a Therapist: Where It Helps and Where It Harms
Structured exercises and rehearsal can help; crisis, delusion and dependency are where it fails.
Opinion · 7 min
Vibe Coding Security Risks: The Check That Was Never Written
AI-generated apps rarely fail through exotic bugs; they fail through omissions nobody asked for.
Opinion · 8 min
What Is Prompt Injection? Why It Is Not Like SQL Injection
The comparison suggests a patch exists; the architecture says otherwise.
Opinion · 9 min
What LLM Guardrails Can and Cannot Stop
A probabilistic filter facing an attacker with unlimited retries is a rate reducer, not a wall.
Opinion · 9 min
Who Is Liable When AI Gets It Wrong?
Blaming the model rarely works; responsibility tends to land on whoever deployed it to the public.
Opinion · 7 min
Will AI Take My Job? Look at Tasks, Not Job Titles
Exposure lives at the level of tasks, and the rungs most at risk are the ones that train experts.
Opinion · 8 min
AI Security, Privacy and Vulnerability Index
Evasion Attacks via Adversarial Perturbations
The failure is confident, which is what removes the signal that would have caught it
Opinion · 4 min
Improper Output Handling
The moment generated code reaches a shell, it stops mattering who wrote it
Opinion · 4 min
Supply Chain Compromises in Open-Source Model Repositories
Copying an identifier into a config is a procurement decision that nobody recorded
Opinion · 4 min
API Parameter Exploitation in Hybrid AI Models
The trusted line is the one nobody drew
Opinion · 4 min
Data Poisoning in Fine-Tuning Pipelines
A few crafted examples bend behaviour on a narrow slice while every aggregate stays flat
Opinion · 3 min
Indirect Prompt Injection in Enterprise Knowledge Bases
The payload arrives through a sanctioned route, sits inert, and fires on somebody else's question
Opinion · 4 min
Model Extraction and Reverse Engineering
A public endpoint is a slow, lossy, complete description of the thing behind it
Opinion · 3 min
Model Poisoning Against Data Poisoning
Same headline, different door, and most defences are guarding the wrong one
Opinion · 4 min
Over-Privileged Autonomous Agents
The credential was broad so the thing would work, and narrowing it now risks breaking a thing that works
Opinion · 3 min
RAG Vector Database Exploitation
The index is filed as infrastructure and behaves as memory, so write access to it is write access to what the system believes
Opinion · 3 min
System Prompt Leakage
Text kept as configuration, which behaves as documentation the moment it escapes
Opinion · 3 min
Continuity
Exit Zero Means Nothing
Six autonomous coding jobs reported success. Three had produced nothing usable. What a green result actually proves, and what it does not.
Verified · Field note · 5 min
Other
Building for the Case Where You Are Compromised
Most systems are designed to prevent a breach. The interesting question is what they are worth after one.
First-person · Founder essay · 4 min
The Model in the Middle
When a language model sits between a person and a system, it inherits both sides' permissions and neither side's judgement
Opinion · 5 min
The Problem I Am Working On
Stated as a problem, because the solution is not mine to describe yet
Opinion · Founder essay · 4 min
The Supply Chain You Did Not Choose
Every dependency you install brings its own dependencies, and the trust decision was made by somebody else
Supported · 4 min
The System Cannot Be Asked Why
Most disputes about automated decisions are really disputes about who has to explain themselves, and to whom
Opinion · 5 min
What a Model Cannot Know About Itself
Asking a system to report its own confidence, its own reasoning, or its own limits produces text, and text is not evidence
Opinion · 4 min
Why I Publish What My Software Cannot Do
Stated limitations are expensive, which is exactly why they are worth reading and worth writing
First-person · Founder essay · 4 min